Upload
vuongthien
View
216
Download
0
Embed Size (px)
Citation preview
19/10/2016
1
A Refresher of the Treatment of Dementia (as we forgot the evidence)
G. Michael AllanEvidence & CPD Program, Alberta College of Family Physicians
Professor, Dept Family Med, U of A.
Faculty/Presenter Disclosure
• Faculty: G Michael Allan
• Relationships with commercial interests over 15 yrs:– Grants/Research Support: None
– Speakers Bureau/Honoraria: None
– Consulting Fees: None
– Other: None
• Where I get Money: – I get pay from Alberta Health and U of A
– I get grants and KT funding from a variety of non-profit sources like ACFP and TOP.
19/10/2016
2
Dementia
1) Understanding dementia treatment trials
2) Options for Treatment
• Cholinesterase Inhibitors Benefit
• Cholinesterase Inhibitors Harms
3) Dose, Severity, withdrawing, prevention.
4) Other options
1) Memantine
2) Cognitive Enhancement.
Understanding Dementia Research
• Dementia category by MSE
– Mild 21‐26; Moderate 10‐20; Severe <10
• Scales! Need to know
– Mini‐mental Status Exam (MSE), 0‐30, higher better
– Alzheimer’s Disease Assessment Scale (ADAS‐Cog), 0‐70, lower better
– Global Impression of Change (physician rated), (out of 7)
• Minimal Important Clinical Difference
– MSE = 1.4 and in ADAS Cog = 4
• To evaluate: What is the mean change & how many got MCID
Placebo Rivastigimine Difference Attain 1.4 change
MSE At 6 months ‐0.6 +0.2 0.74 ?
19/10/2016
3
What’s NEW ?
MSE change(out of 30)
ADAS‐Cog change (out of 70)
Global Impression of Change
2009 (9 RCTs) 0.82 1.99 0.66 (0.55‐0.79)
2015 (13 RCTs) 0.74 1.79 0.68 (0.58‐0.80)
2016
2006
Rivastigmine Cochrane reviews 2009 and 2015
Therapies for Dementia
Acetylcholinesteraseinhibitors
Memantine?
19/10/2016
4
Count along
11 Vitamin B612 Vitamin B1213 Lecithin14 Ibuprofen15 Acetyl-L-carnitine16 Statins17 Nicergoline18 Hydergine19 Propentofylline20 Nimodipine
31 Cannabinoids32 Omega 3 (prevent or treat)33 Melatonin34 Folic acid35 Vinpocetine (herbal)36 Propentofylline37 Nimodipine38 Huperzine A39 Validation therapy40 Aroma therapy for dementia
1 Ginseng2 Ginkgo biloba3 Piracetam4 Physostigmine5 Nicotine6 Selegiline7 Vitamin E8 Indomethacin9 Thiamine10 D-cycloserine
21 Huperzine A22 Velnacrine23 Procaine24 Metal Protein Attenuating
compounds25 Carbohydrates26 Cerebrolysin27 ASA28 Alpha lipoic acid29 Clioquinol30 Almitrine-Raubasine
41 Homeopathy42 Snoezelen (stimul. therapy)43 Physical Exercise44 Reminiscence Therapy45 Music Therapy46 Respite Care47 Light therapy48 TENS49 HRT50 Metrifonate51 Naftidrofuryl52 BP lowering53 Latrepirdine54 More,…
Meta‐analysis of Dementia RCTs
• What is the scientific evidence for Cholinesterase Inhibitors in the treatment of Alzheimer's disease.
• 22 Trials: 12 Donepezil, 5 Rivastigmine, 5 Galantamine: 27 to 978 pt/trial, 6 wks‐3yrs long
• Findings: 1.5‐3.9 (ADAS‐cog & Min clinical sign ≥ 4)
• Limitations: Numerous
– ITT flaws (pt exclusion after randomization)= 15/22 (68%),
– Last Observation Carried Forward (declining illness)
– Use of Means (in scales),
– No correction for multiple comparison
– Funding (often authored by employees)
BMJ 2005; 331: 321‐27
19/10/2016
5
200 Dementia Patients
100 Take PlaceboMean MSE=20
100 Take DonepezilMean MSE=20
1 Year TrialAdverse Events Intolerable
10 withdrawal wk 1
100 Taking PlaceboMean MSE=16
90 Taking DonepezilMean MSE=16
10 withdrawalsMean MSE=20
Final 100 DonepezilMean MSE=16.4
Cholinesterase Inhibitors: Summary
• Cholinesterase trials vs Placebo: MSE 10‐26*– Poor reporting (e.g. 12% of Donepezil report mortality)
– Alzheimer’s Disease Assessment Scale (ADAS‐cog)=4 is clinical significant,
– Quality of Life scores unchanged
* Health Technol Assess 2012;16(21). 1) Cochrane. 2006;(1):CD001190 (10mg x 6 months). 2) Cochrane 2006; 1: CD001747. 3) Cochrane 2015; 9: CD001191. ( *2009) 4) Cochrane 2006; Issue 1: CD005593.
Donepezil1 Galantamine2 Rivastigmine3 All4
MMSE 1.44 ? 0.74 1.37
ADAS - Cog 2.81 3.38 1.79 2.73
ADAS – Cog of 4 ? NNT 6 NNT 18* ?
Glob Clin State NNT 10 NNT 7 NNT 13 NNT 14
Adverse Events NNH 18 ? NNH 8 NNH 8
? Not given
19/10/2016
6
Other Outcomes: Example Donepezil
• ADL & IADL: Most statistically significant– Lots of different ones used, so summing up hard
– Basically, move about <4% on different scales.• E.g. updated rivastigmine: change = 2.15 out of 100.
• Quality of Life: Patient rated. – No difference.
• Behavior: Primarily NPI used– No difference 12 wks, 10mg
– Difference (24 wks, 10 mg): 2.94 (out of 144)
Cochrane 2006;1:CD001190. Ann Intern Med. 2008;148:379‐97. Cochrane 2015,(9): CD001191.
Adverse Events: Example Donepezil
• Statistically significant
– Anorexia: 7.3% vs 2.1%, NNH 20
– Diarrhea: 14.5% vs 5.3%, NNH 11
– Nausea: 14.5% vs 5.4%, NNH 11
– Vomiting: 11.3% vs 4.7%, NNH 16
– Weight Loss: 8.2% vs 4.5%, NNH 28
– Fatigue: 9.4% vs 4%, NNH 19
– Asthenia (weakness): 7.9% vs 4.7%, NNH 32
– Dizziness: 8.1% vs 5.4%, NNH 38
– Insomnia: 9.9% vs 4.4%, NNH 19
• Others Borderline (accidental injury, rhinitis)
Ann Intern Med. 2008;148:379‐397. Adverse Event Appendix Table 4
19/10/2016
7
Is one better than another?
• 3 Trials compare Head to Head1
– Multiple Flaws & potentially biased• Industry funded, Employee written, results favoring sponsor.
– In Meta‐analysis : “There is no evidence of any difference between them”2
• Four new RCTs: 3 weak and no reliable difference.3
– Fourth: Rivastigmine vs donepezil,• No difference in cognition/behavior
• Marginal (very) differences in function and global effect.
• Bottom‐Line: No reliable difference.
1) Lancet Neurol 2004; 3: 622:26. Therapeutics Letter 2005; 56:1‐4. 2) Cochrane Database Syst Rev. 2006 Jan 25;(1):CD005593. 3) Health Technol Assess 2012;16(21).
Prevention of Dementia:
• Vitamin E : No help
• Exercise: No help.
• Meta‐analysis Donepezil:
– In 1 of 2 trials, 1 of 5 scores had a 3% less decline
– Stopping due to adverse events: NNH 7.
• Meta‐analysis Galantamine:
– Marginal to no clinical Benefit
– ++ Harms: NNH (for death) = 94.
• Bottom‐Line: None.
1) NEJM 2005; 352:2379‐88. 2) Cochrane 2015;(4):CD005381 3) Cochrane Database Syst Rev. 2006;3:CD006104. 4) Cochrane 2006;(1):CD001747. Therapeutics Letter 2005; 56:1‐4.
19/10/2016
8
What about withdrawing?
• 295 Community dwelling Patients on Donepezil (most >2 yrs)– mean age 77, mean MSE 9, followed 1 yr.
– Stopping of med worsened MSE by 1.9 pts• Less effect if severe dementia (<9 MSE)
• Don’t give number attaining MCID (1.4)
– Withdrawal from study more if stopped!
– Death: no difference
N Engl J Med 2012;366:893-903.
Does Dose Matter?• Moderate dose may matter
– Donepezil: ADAS‐Cog 2.15 (5mg) vs 2.45 (10mg)• MSE results equal at 5 or 10mg.
– Rivastigmine: ADAS‐Cog 0.84 (1‐4 mg) vs 1.99 (6‐12mg)
– Galantamine: Global rating OR 1.17 (8mg) vs 1.63‐1.84 (16‐36mg)
• Bottom‐Line: Use mostly indirect comparison. May be that some low doses have a somewhat weaker effect. Probably avoid Galantamine 8mg and rivastigmine 1‐4 mg.
1) Cochrane. 2006;(1):CD001190 (10mg x 6 months). 2) Cochrane 2006; 1: CD001747. 3) Cochrane 2015; 9: CD001191. ( *2009)
19/10/2016
9
Does Severity Matter
• Bottom‐Line: Maybe more effective in moderate severity, but no formal patient level analysis, data sparse and somewhat inconsistent?
Donepezil @ 24wks Change in MSE
Donepezil @ 24wks Varying Cog Outcomes
Donepezil @ 24wks Change in Function
Donepezil @ 24wks Global Change
XHealth Technol Assess 2012;16(21).
Cognitive Enhancement
• 15 RCTs with 718 participants– Example: naming objects & people, word association, remembering the past, discussion of hobbies, activities & current affairs, using money, knowing the way around and orientation topics
• Outcomes: change in scale– ADAS‐Cog at 1‐12 months: 2.27 (0.99, 3.55)
– MSE at 1‐12 months: 1.74 (1.13, 2.36) better• Longer seems better but based on only one study.
• Bottom‐Line: Seems to have some positive effects (similar to drugs).
Cochrane 2012; 2: CD005562.
19/10/2016
10
Drugs with Potential: Memantine
• Mostly Moderate - Severe Dementia– ADCS -ADL score, Severe impairment battery,
Functional assessment Staging, Clinician Impression of Change (CIBIC): All 0-4% change
– Possibly <agitation (NNT= 63) - if already on– Well Tolerated (no diff in drop-out due to AE)– Other studies use SMD statistic & can’t interpret.3
• Bottom-Line: Effects are small & inconsistent.
Cochrane 2006;(2):CD003154. Health Technol Assess 2012;16(21). 3) PLoS ONE 10(4): e0123289.
Combining Medicines
• Adding Memantine to Cholinesterase inhibitors
– 4 RCTs with 1439 pts, ~10 (mean range 7‐16)
• 6 months
– ADAS cog: 1.6 better
– MSE: 0.5 better
– Neuro‐Psychiatric inventory: 1.6 better (out of 144)
• Bottom‐Line: The best this combination can offer is 1‐2% change.
Health Technol Assess 2012;16(21).
19/10/2016
11
A New Hope: Aducanumab
Nature. 2016 Aug 31;537(7618):50‐6.
PET Scans of Amyloid Deposits.
A New Hope: Aducanumab
• RCT 165 pts, age 73, 50% female, dose q‐month IV 1,3,6,10 mg/kg or placebo
• Outcome: at one year
• Bottom‐Line: Preliminary evidence suggests potential benefit but very early and will need evidence in moderate to severe dementia.
Measure Scale Worse Baseline Placebo vs 10 Main Types of AE
MSE 0‐30 low 24 2.7 vs 0.5 worse Edema (NNH 3), Headache (NNH 5), superficial iron deposition in CNS (8)
CDR‐SoB 0‐18 high 3.2 1.8 vs 0.7 worse
FCSRT 0‐48 lower 14 No diff
Discontinue due to AE 10% vs 31% (ss)
Clinical Dementia Rating—Sum of Boxes. Free and Cued Selective Reminding Test;
Nature. 2016 Aug 31;537(7618):50‐6.
19/10/2016
12
Summing‐Up
• Cholinesterase inhibitors basically– Improve dementia for 1 in 10 (over placebo)
– Cause Adverse events severe enough to stop taking for 1 in 10 (over placebo)
– If really working (or was working), stay on.
• All meds the same, Don’t give for prevention, don’t combine meds
• Cognitive enhancement measures wherever possible
• Hope for Aducanumab (but still just hope)
Getting a Complete Drug History
Why are you on the “white one”?
It’s for my dizziness.
How long have you been dizzy?
A while
How long’s a while?
Sometime now
Do you know when it started?
My Sisters birthday.
When is your sister’s birthday?
Which sister?
19/10/2016
13
Donepezil: Stay Safe